research methods
process
How did we traverse the space of music listening and the future of music in a digital age?
Overview

There was no easy way to begin the process of exploring, understanding, and experimenting in the music streaming space. With so many opportunities and experiences to cover, our team decided to employ a large range of both primary and secondary research methods as we tackled the question of “what is the future of music streaming and listening?” 

The following chronological sections feature a brief summary for each method and the most important results to have been produced from them. For those who are interested in a more thorough walkthrough of our research process, an academic research paper that covers the same material but in more detail can be found at this link.

expert-interviews

Music is a fundamental human experience. You find it across all ages, in all places, across time & history. We all listen to music, but when our team sat down together back in January and tried to tackle this area, it was really hard to put into words, “what is music, and where is it headed?” So, as we began delving into the black box of music listening and the problem space as a whole, we defined a few focuses for ourselves. 

The first focus was understanding the future of music. We were lucky enough to be sitting in the backyard of Carnegie Mellon University, and it didn’t take long before we found academics who were experts in fields ranging from user-sensing systems, to conversation, to smart-agents and those who had already researched music in their careers.

Expert Interviews
6

John Zimmerman, Mayank Goel, Dr. Jill Lehman, Michal Luria, Paul Pangaro,
Nik Martelaro

Context-Awareness
1

Contextual-awareness is key to the future of music, and predicts the ability to create a frictionless experience in music listening

Multi-modal Experiences
2

Increasing accessibility of smart devices opens up doors to multi-modal experiences that will bolster dynamic and holistic listening.

Direct Channels Between Artists and Listeners
3

There is an opportunity to develop novel channels of communication between artists and their listener fans within this transition to a digital streaming platform age.


With these first few insights in mind, we came across another question. What is a possible future with music? At this point in the project, we had a pretotype assignment that got us really excited to jump into making with our hands, failing fast, learning from doing. And that’s exactly what we did.

The Cove / Pretotype

Around this time, we shared a conversation with our client about how they’ve been exploring a “Bartender” personality to Alexa. And the personalities to CUIs was something that’d already come up in previous expert-interviews with academics like John Zimmerman and Dr. Jill Lehman. So, this got us thinking...what if your Alexa was less Alfred Pennyworth, and more Sam Malone? We decided that we also wanted to explore this more informal personality, one that reaches informality because you’ve built a relationship together over time. To test it out, we embedded this friendlier, open, conversive personality into our pretotype.


Meet, The Cove. We sectioned off a space right next to our room in the lab, hung up shower curtain from Target, and started putting people in The Cove. It was this future state of music, where they got to interact with a formal and informal CUI and pick out music they wanted to listen to.

The context-aware sweet spot


What we found was an emerging context-awareness sweet spot for users and this depended heavily on the current-activity and mood of our users. We asked users about their preferences for dialogue length and formality, and there were repeated mentions of using the informal/chatty CUI next to their friends, while being with your parents often led to a preference for a formal CUI.

Context-Awareness
1

Context and current-activity are most influential in determining users' preference in formality.

Perception of CUIs
2

Satisfaction was tied to users' perception of the CUI's capabilities, and how much it differed from real capabilities.


Beyond testing music recommendation, we wanted to see if CUIs could be used to strengthen the artist-listener relationships we had uncovered earlier, like sharing “nuggets” of information about interesting stories behind songs.

Nuanced information
1

Users with existing deep connections to artists prefer information that is obscure of "Easter-eggy" in nature.

Cultural Capital
2

There is perceived value in learning about cultural figures in order to better engage in social conversations later.

guerrilla research

At this point we had started building a better understanding of music and the future of music as a space, but what about the users. They are a key consideration in this project and our team knew that we wanted users to be the main focus of our service.

We wanted to answer this question “how do users feel about music streaming today.” And we searched for the answers by heading out onto the streets of Pittsburgh to conduct some guerrilla research.

Participants
20

late-teens to late-sixties,
varying occupations,
varying socio-economic backgrounds

Hang the Algorithm
1

In part due to lack of trust and control in algorithms, users want to bring back the "human touch" once present in DJ sets and live shows.

Tangible Music Algorithm
2

Lack of transparency in algorithm recommendations leaves users confused and dissatisfied with current streaming platforms.

"I can't own a streamed song"
3

There is an opportunity to redefine and recreate a sense of ownership that was lost in the transition from physical music artifacts to a digital streaming age.

Community in Music
4

There is a desire for listener-to-listener, fan-to-fan communities, as music listening is not just an insular experience in real life.

Eras in Music Listening
5

Music is often strongly associated with phases and memories of users' lives, and there is an opportunity to capture this in streaming.



At this point in our research, we hit late February. In 2 months, we collected all these insights about the future of music and how users feel about streaming today… but this still weren’t capturing the whole picture of music listening. We missed the day-to-day role of music in our lives and the needs that emerge from this space.

So, it’s less about just listening to music, and it’s more about how music is used in-real-life. Our research efforts from this point forward followed this pivot in focus, and dived deep into the IRL experience of music listening.



Diary Study + Directed Story telling Uncovered IRL Music Listening

With our new direction in IRL focus, we set about exploring what does IRL music listening look today. For that, we really had to understand fundamentally how people use or did not use, music throughout their day. We accomplished this by doing four diary studies with followup interviews and nine directed story telling interview sessions. What we found was rather surprising but also somewhat obvious, that music is multi faceted tool in IRL.


Music as an environmental tool

“I use music to feel like I’m not on the bus surrounded by people.”

Participants mentioned that music served as a tool to change their perception of their environment. Users used music  to redefine their work environment, create space, and to set a background mood for events in their day.

Music as a productivity tool

“Music is my Adderall.”

Participants mentioned circumstances where music increased their focus, creativity, and ability to succeed. It helped them to complete chores, work, spreadsheets, etc.

Music as a emotional tool

“If I’m going out later, I’m gonna listen to a ‘let’s get f$$ked up’ song.” 

Music and emotion are very closely bundled. Participants used this relationship to alter their mood, to relive the past, and to augment their mood during IRL experiences.

Music as a social tool

“I introduce my music to my children. It feels like a new connection.” 

Participants mentioned how they used music as a tool for social cohesion. Whether it was curating music for another at work, sharing a playlist, or discussing the latest Kanye album, it was obvious that music is used as a social tool in IRL.




Customer Journey Mapping and Speed dating uncovered user IRL Music Listening Friction

Distilling our diary studies and directed story telling led us to uncover very specific areas where users struggled with IRL music streaming. To validate these areas - we decided to ideate and storyboard 15 possible music futures. We then conducted 9 separate storyboarding sessions with a variety of participants and found some unanimous areas of need in IRL listening.

Selecting Music for Others in Social Situations
1

“I never know what to play for other people… that gives me anxiety.” 

Selecting music for others in social situations was by far the most mentioned place of friction. Curating for others is an intimate activity and user’s would love to see ways to help ease this activity. Especially for situations that are more formal vs. informal, are heightened due to social significance, and revolve around individuals that are not familiar with.

Having to pick music for yourself
2

“Sometimes I just get tired of selecting what to listen to next - that's why I like the radio.”

Another area that was mentioned frequently was the trouble of choosing what to listen to. With an almost infinite selection of music on streaming apps today -  just choosing a song, a playlist, or even an album appears to be a major pain point. This leads them to choose lazily and sometimes ending up in a deeper filter bubble. Solutions do exist for some - user’s like using a Dj’ed radio as a way out - or artist playlists. Yet even choosing to enter these modes can be burdensome

Streaming is not context-aware
3

“I don’t want slow songs at the gym… so when they come up I have to skip them.”

‍Lastly a big problem area that was frequently mentioned was the lack of streaming experiences being context aware. What we found is that problems arise in the music stream experience because applications today are: Not aware of the activity you are doing, where you are doing said activity, and why you are doing it.User’s want systems to be better aware of their context; yet within boundaries of the uncanny valley and privacy

Another aspect of this context awareness stems to a more  device problem. Users do not want to be constantly switching between devices. Friction that arises from transferring playback between the many touch points in IRL listening.While it is still better than it’s ever been - users wish this more frictionless and automatic.

Prototyping helped us uncover possible futures for IRL streaming exerpiences

Moving away from areas of opportunity and friction - we decided to ideate on what some possible futures of music streaming could look like. Our goal was to figure out where our efforts most aligned with user's IRL needs and how they responded to more tangible prototypes

Prototypes
9

33 participants
640 minutes of interviewing
20+ insights and opportunities

Narrowing down our Oppertunity space

Using the prototypes of future IRL experiences as a jumping off and narrowing tool, we where able to see three clear directions that we could swing the project towards. Click through our opportunity spaces to see where we landed!